I want to copy a small 4 x 4 matrix from host to device. My host array/matrix is defined as follows:
Can I do something like as follows:
device constant float d_m;
and to copy use this in host:
CUDA_SAFE_CALL(cudaMemcpyToSymbol(d_m, m, 16 * sizeof(float));
Is this the correct way of doing this? I am currently using a non-cuda machine to code CUDa software(!) and am unable to test this and would be grateful if someone could correct me if I am wrong!